An approach to the regional forecast for zones of oil inflow from Bazhen-Abalak formation has been formalized and tested. The task was to classify the spatial attributes by machine learning through precedents by algorithm of single decision tree with the genetic selection of combination of such attributes. The rules have been retrieved and the factors have been identified which influence the forecast for zones of oil inflow from Bazhen-Abalak formation intervals. The results are shown in the regional forecast scheme with identification of Bazhen-Abalak formation sweet spots in KhMAO-Yugra region of Russia. Such sweet spots can be correlated with perspective zones to get the inflow from the Bazhen-Abalak formation.
References
1. Brekhuntsov A.M., Nesterov I.I. Jr., Nechiporuk L.A., Bituminoznye glinistye otlozheniya
bazhenovskogo gorizonta – prioritetnyy strategicheskiy ob"ekt
neftedobychi v Zapadnoy Sibiri (Bituminous clay deposits of Bazhenov horizon
– priority strategic facility of oil production in Western Siberia), URL:
http://oilgasjournal.ru/vol_10/brekhuntsov.pdf
2. Breiman L., Friedman J., Olshen R., Stone C., Classification and regression
trees, Wadsworth & Brooks, Pacific Grove, CA, 1984, 368 р.
3. Hastie T., Tibshirani R., Friedman J., The elements of statistical learning: data
mining, inference, and prediction, 2009, 739 p.
4. Quinlan R., Programs for machine learning, San Mateo: Morgan Kaufmann,
1993, 302 p.
5. Poli R., Langdon W.B., McPhee N.F., A field guide to genetic programming,
2008, 250 p.
6. Standardized Verification System (SVS) for Long-Range Forecasts (LRF),
New attachment II-9 to the Manual on the GDPS (WMO-No. 485), 2002, V. I,
pp. 14–20.
